What is the log used for?

To persist a complete, auditable record of a finished phase.

How is the next phase number decided?

By counting existing phase-*.md files and zero-padding to 3 digits.

How does it prevent duplicate logging?

It writes last-log-state.json with the transcript hash and turn count to stop auto-logger re-logging.

log

npx machina-cli add skill lemon-etvibe/prompt-vault/log --openclaw

Files (1)

SKILL.md

1.8 KB

Record completed work from the current conversation session as a phase log.

Procedure

Verify .local/logs/ directory exists (create if missing)
Determine next phase number from existing phase-*.md file count (3-digit zero-padded)

Write phase-NNN.md in the following format:

# Phase NNN: $ARGUMENTS (or auto-inferred title)

- **Date**: YYYY-MM-DD
- **Session**: (session ID or timestamp)

## User Prompt
> Record the user's original request as close to verbatim as possible

## Actions
- List major actions performed by Claude in chronological order
- Include tools used, files read, files created/modified

## Results
- Summary of deliverables (generated file paths, key findings)
- Key data or metrics

## Decisions
- Decisions made and their rationale

## Next
- Next steps or open items

Update .local/logs/_index.md: add/update the phase row

| NNN | Title | done | YYYY-MM-DD | One-line summary |

Update auto-log state file (.local/logs/last-log-state.json) to prevent auto-logger duplication:
- Compute transcript hash via Bash:
```
shasum -a 256 ~/.claude/projects/.../SESSION_ID.jsonl | cut -d' ' -f1
```
  Prefix with sha256: to form "sha256:<hex>"
- Write state:
```
{
  "lastLogTimestamp": <current_timestamp_ms>,
  "lastTranscriptHash": "sha256:<hash>",
  "lastLogTurnCount": <total_turn_count>,
  "lastPhaseNumber": "<NNN>",
  "lastTrigger": "manual"
}
```
- This ensures the auto-logger (Stop/PreCompact hooks) won't re-log the same content
Output completion message

Source

git clone https://github.com/lemon-etvibe/prompt-vault/blob/main/skills/log/SKILL.mdView on GitHub

Overview

This skill creates a structured phase log whenever a phase completes. It writes phase-NNN.md with date, session, user prompt, actions, results, decisions, and next steps, then updates the index and a state file to avoid duplicate auto-logging.

How This Skill Works

It verifies the local logs directory exists, counts the phase-*.md files to determine the next phase number, and writes phase-NNN.md in a standard format. It also updates the _index.md and last-log-state.json to support auditing and deduplication.

When to Use It

After finishing a phase's dialogue and deliverables, to archive the work
When a phase concludes and requires a traceable record for audit or handoff
Before concluding a session to provide a persistent summary for review
When a multi-step task spans phases and you need phase-level separation
To prepare a handoff packet with the next steps and decisions

Quick Start

Step 1: Ensure the .local/logs/ directory exists (create if missing)
Step 2: Count phase-*.md to compute the next NNN (zero-padded to 3 digits)
Step 3: Write phase-NNN.md with the required sections, then update _index.md and last-log-state.json

Best Practices

Always ensure the .local/logs directory exists before writing
Include the user's original request verbatim in the User Prompt section
Number phases with 3-digit NNN (phase-001, phase-002, ...)
List major actions and tools used in chronological order
Keep the log file content readable and consistent for audits

Example Use Cases

Phase 001: Onboarding request captured with date, session, and user prompt
Phase 002: Data extraction and transformation with tools listed
Phase 003: Report generation and file paths documented
Phase 004: Tool orchestration and file reads/writes recorded
Phase 005: Final review and handoff prepared

Frequently Asked Questions

Add this skill to your agents